Conversational AI Engineer (Game Prototype)

Location: Remote; at least 4 hours of overlap with US Pacific required

Company

hirify.global is prototyping a game centered on AI-driven conversations with non-player characters.

What you will do

• Design and build an end-to-end voice conversation pipeline between players and NPCs.
• Evaluate and integrate speech-to-text, language model, text-to-speech, and speech-to-speech technologies.
• Develop approaches for conversational memory, context, personality, and NPC behavior.
• Prototype real-time interactions and optimize latency, quality, reliability, and cost.
• Choose which components run on-device or in the cloud, measure response time, cost per conversation minute, and failure rates.
• Build and integrate the system in Unity and C#, creating an architecture that can evolve into production.

Requirements

• 5+ years of professional software engineering experience.
• At least one shipped project built in Unity with C#, such as a released game, application, or production tool.
• A working end-to-end voice conversation loop built independently, covering speech input, language model processing, and speech output.
• Hands-on engineering experience building playable prototypes rather than focusing primarily on prompt writing, single-API chat applications, research, or model training.
• Availability for remote work with at least 4 hours of overlap with US Pacific time.

Nice to have

• Game AI, NPC systems, or interactive entertainment experience.
• Conversational memory and agent architecture experience.
• Experience with on-device speech or language models.
• Experience building multilingual or language-focused AI systems.

Culture & Benefits

• Remote contractor engagement.
• Initial 3-month project focused on designing and building a working conversational AI prototype in Unity.
• High ownership over technology selection, architecture, and implementation.
• Potential to continue with the team after the prototype stage.

Hiring process

• Submit a resume and cover letter, a link or demo of a self-built voice or conversational system, and short technical answers.
• Shortlisted candidates will present their demo during a 30-minute call.